lable at ScienceDirect

Quaternary Science Reviews 36 (2012) 139e153

Contents lists avai

Quaternary Science Reviews

journal homepage: www.elsevier .com/locate/quascirev

Northern Iberian abrupt climate change dynamics during the last glacial cycle:A view from lacustrine sediments

Ana Moreno a,*, Penélope González-Sampériz a, Mario Morellón a,b, Blas L. Valero-Garcés a,William J. Fletcher c

aDepartment of Geoenvironmental Processes and Global Change, Pyrenean Institute of Ecology e CSIC, Avda. Montañana 1005, 50059 Zaragoza, SpainbDepartment of Geographical and Earth Sciences, University of Glasgow, East Quadrangle, University Avenue, Glasgow G128QQ, United Kingdomc Institute of Geosciences, Goethe-Universität, Frankfurt am Main 60438, Germany

a r t i c l e i n f o

Article history:Received 9 March 2010Received in revised form18 May 2010Accepted 11 June 2010Available online 14 July 2010

* Corresponding author.E-mail addresses: amoreno@ipe.csic.es (A. More

González-Sampériz), mariomm@ipe.csic.es, marioMorellón), blas@ipe.csic.es (B.L. Valero-Garcés), w(W.J. Fletcher).

0277-3791/$ e see front matter � 2010 Elsevier Ltd.doi:10.1016/j.quascirev.2010.06.031

a b s t r a c t

We present a palaeoclimatic reconstruction of the last glacial cycle in Iberia (ca. 120,000e11,600 cal yrsBP) based on multi-proxy reconstructions from lake sediments with robust chronologies, and witha particular focus on abrupt climate changes. The selected lake sequences provide an integrated approachfrom northern Iberia exploring temperature conditions, humidity variations and land-sea comparisonsduring the most relevant climate transitions of the last glacial period. Thus, we present evidence thatdemonstrates: (i) cold but relatively humid conditions during the transition from MIS 5 to MIS 4, whichprevailed until ca. 60,000 cal yrs BP in northern Iberia; (ii) a general tendency towards greater aridityduring MIS 4 and MIS 3 (ca 60,000 to 23,500 cal yrs BP) punctuated by abrupt climate changes related toHeinrich Events (HE), (iii) a complex, highly variable climate during MIS 2 (23,500 to 14,600 cal yrs BP)with the “Mystery Interval” (MI: 18,500 to 14,600 cal yrs BP) and not the global Last Glacial Maximum(LGM: 23,000 to 19,000 cal yrs BP) as the coldest and most arid period. The last glacial transition starts insynchrony with Greenland ice records at 14,600 cal yrs BP but the temperature increase was not soabrupt in the Iberian records and the highest humidity was attained during the Allerød (GI-1a to GI-1c)and not during the Bølling (GI-1e) period. The Younger Dryas event (GS-1) is discernible in northernIberian lake records as a cold and dry interval, although Iberian vegetation records present a geograph-ically variable signal for this interval, perhaps related to vegetation resilience.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

The last glacial cycle (ca. 120,000e11,600 cal yrs BP) wasa dynamic period when rapid climate changes, called Dansgaard/Oeschger (D/O) cycles and characterized by abrupt warming andgradual cooling, occurred with a periodicity of ca. 1450 years (Wolffet al., 2010). Understanding the response of different ecosystems tothese rapid climatic events is of special interest in the context ofpresent-day global warming but, unfortunately, the mechanismbehind rapid climate oscillations, the teleconnections that transferthe signal all around the globe, and the impacts of rapid climatechanges on terrestrial and marine ecosystems are still far frombeing totally understood (Broecker, 2000). In fact, it is known that

no), pgonzal@ipe.csic.es (P..morellon@ges.gla.ac.uk (M..fletcher@em.uni-frankfurt.de

All rights reserved.

some of the climate events of the last glacial cycle were notsynchronous, such as the timing for the maximum glacier advanceat different latitudes (Hughes and Woodward, 2008; Clark et al.,2009), but the causes remain unexplained. In particular, the lastglacial-interglacial transition (LGIT, 15,000e9000 cal yrs BP) hasa special interest since many processes and components of theclimate systemwere involved in a total restructuring of the climateat a global scale. That transition occurred in several steps, some ofthem still poorly known in terms of their hydrological signal orinternal structure, such as the Mystery Interval (MI) (17.5e14.5 calkyr BP) (Denton et al., 2006). To address all these questions, it isnecessary to assess the synchrony or asynchrony between differentrecords from different archives, and this is one of the foci of INTI-MATE (INTegration of Ice-core, MArine and TEerrestrial records)group (Hoek et al., 2008).

The Iberian Peninsula (IP) constitutes a key location foranswering questions related to the transference of the climatesignal from high- to mid-latitudes. The IP is an especially sensitiveregion to climate changes due to its location at geographical

A. Moreno et al. / Quaternary Science Reviews 36 (2012) 139e153140

(subpolar versus subtropical latitudes) and atmospheric (westerlywinds versus north-African influences) boundaries (Moreno et al.,2005; Bout-Roumazeilles et al., 2007). In addition, its locationleads to the expression of some of the “cold northern events”during last glacial cycle as “dry southern events”, as inferred fromdust accumulation (Moreno et al., 2002) and pollen composition inmarine cores surrounding the IP (Sánchez-Goñi et al., 2002;Fletcher et al., 2010). It remains necessary to evaluate the precisespatiotemporal nature of terrestrial ecosystem change, as sug-gested by recent lake (González-Sampériz et al., 2006) and spe-leothem records (Moreno et al., 2010). Understanding the effects ofpast abrupt climate changes may help to predict and minimize theimpact of future global warming (Costanza et al., 2007) in the IP,one of the most vulnerable areas in the context of the Mediterra-nean region (Solomon et al., 2007).

Iberian terrestrial records, supported by the study of terrestrialtracers (pollen) in marine cores, have allowed the characterizationof the response on land to climate change and the discrimination oflocal or regional signatures, both necessary tasks to complete andimprove the palaeoclimate reconstructions carried out in Europeduring the last glacial cycle (e.g., Wohlfarth et al., 2008). Addi-tionally, lakes are systems where changes in water availability canbe recorded in the sediments in a more direct way than tempera-ture variations (e.g., Cohen, 2003). Thus, the integration of severalproxies (physical properties, sedimentary facies, geochemicalcomposition, diatom and pollen assemblages, etc.), can lead to thereconstruction of past lake levels, and thus to the estimation ofprecipitation-evaporation balance (e.g., Morellón et al., 2009a).Furthermore, other environmental changes such as vegetationcover and land use can be inferred from palynological studies(Morellón et al., in press; Rull et al., in press). Lake sediments canoften provide continuous, high-resolution records with robustchronologies, thus providing detailed and comprehensive palae-oenvironmental reconstructions.

The study of Iberian Quaternary lake sequences with the aim ofreconstructing palaeoclimatic or palaeoenvironmental conditions isrooted in the long history of sedimentological studies of pre-

Table 1Lake records from the IP reviewed in this paper.

Coordinates Lake type

Northern Iberia1. Enol Lake 43�110N; 4�090W; 1070 m asl Karstic e glacial

2. Comella Hollow 43�160N; 4�590W; 850 m asl Karstic e glacial

West-Northwestern Iberia3. Sanabria Lake 42�070N; 6�420W; 1000 m asl Glacial

Iberian range and Central Iberia4. Laguna Grande 42�020N; 3�010W; 1500 m asl Glacial5. Fuentillejo maar 38�560N; 4�30W; 635 m asl Volcanic

6. Laguna del Hornillo 41�580N 2�00W Glacial

Pyrenees and Northeastern Iberia7. Banyoles Lake 42�070N; 2�450E; 173 m asl Karstic

8. Villarquemado palaeolake 40�300N; 1�180W; 987 m asl Tectonicdepression

9. El Portalet peatbog 42�480N; 0�230W; 1980 m asl Glacial10. Saladas de Monegros

(Salineta, La Playa, Mediana.)41�280N; 0�090W; 350 m asl Dissolution and

aeolian deflaction11. Ibón de Tramacastilla 42�430N 0�230W; 1640 m asl Glacial

12. Estanya Lake 42�020N; 0�320E; 670 m asl Karstic

Proxies: PHYS, physical properties; SED, sedimentological description; GEO, geochemistreconstruction.

Quaternary formations (Cabrera and Anadón, 2003; Valero-Garcés,2003). However, only recently and thanks partly to new technicalimprovements (both in the field and laboratory) and to theconsolidation of new Spanish research groups, has climate recon-struction been tackled using a multi-proxy strategy and robustchronological frameworks. Thus, the number of palaeoclimatestudies from lake records in the IP hasmarkedly increased as well asthe quality of the records, in terms of their continuity, chronologicalaccuracy, effective temporal resolution and the range of analyticalmethods combined (Valero-Garcés and Moreno, in press). Weconsider a review of the key published data timely because, sincelake response to climate is non-linear, it is critical to synthesize largedata sets to distinguish clearly local influences from broad-scaleregional patterns (Fritz, 2008). In addition, we highlight the mostcritical gaps in the information (in terms of both spatial andtemporal coverage) to help plan future research in the IP.

2. Study sites

The purpose of this paper is not an exhaustive compilation oflast glacial Iberian lake records but a summary of the mostrecent work that fulfills the following requisites: (1) the palae-oclimate interpretations are based on multi-proxy reconstruc-tions from lake sediments, including sedimentologicaldescription and physical or geochemical data from the lacus-trine sequences and not only palynological data as occurs in thecase of many well-known studies, and (2) the chronology isindependent, robust and accurate, based on calibrated AMS14C dates, UeTh dating or Optically Stimulated Luminiscence(OSL), if applicable. With the selected records, this study aims tocarry out a regional palaeoclimate synthesis (Table 1, Fig. 1)covering the last glacial cycle, since last glacial inception (about120,000 cal yrs BP) to the onset of the Holocene (11,600 cal yrsBP). Up to now, none of the available climate reconstructionsfrom southern IP lake records spanning the last glacial anddeglaciation intervals is based on a multi-proxy strategy. Thus,Padul peatbog from southeast IP is only based on pollen data for

Proxies Chronology References

PHYS, SED, GEO,BIO, POL

38e2.5 cal kyr BP Moreno et al. (in press-a)

SED, GEO Base at 42 cal kyr BP Jiménez Sánchez andFarias (2002)

PHYS, SED, GEO, BIO 25e0 cal kyr BP Rico et al. (2007)

SED 20e0 cal kyr BP Vegas (2007)SED, GEO, POL 700e0 cal kyr BP Vegas et al. (in press) for last

50 cal kyr BPSED 27e0 cal kyr BP Vegas (2006)

SED, GEO, POL 30e5 cal kyr BP Pérez-Obiol and Julià (1994),Valero-Garcés et al. (1998)

SED, GEO 120e0 cal kyr BP Valero-Garcés et al. (2007)

SED, GEO, POL 33e5 cal kyr BP González-Sampériz et al. (2006)SED, GEO, POL ca 20e0 cal kyr BP Gonzalez-Samperiz et al. (2008)

and references thereinSED, GEO, POL 30e0 cal kyr BP García-Ruíz et al. (2003),

Montserrat (1992)PHYS, SED, GEO, BIO 20e0 cal kyr BP Morellón et al. (in press,

2009a)

ry; BIO, biological indicators (diatoms, ostracods, quironomids); POL, palynological

Fig. 1. Outline map of mainland Spain and the Balearic Islands showing the broad division into “Variscan” (pink) and “Alpine” (green) Spain and the Cenozoic basins (light yellow)(modified from Gibbons and Moreno, 2002; Vera, 2004). Lake sites considered in this study are indicated by black circles (see also Table 1) while black squares mark the position ofother sites cited in the text (marine, speleothem and pollen sequences). (For interpretation of the references to colour in this figure legend, the reader is referred to the web versionof this article.)

A. Moreno et al. / Quaternary Science Reviews 36 (2012) 139e153 141

the glacial interval (Pons and Reille, 1988) and the chronologyfor this interval is not well constrained in the new 107-m longborehole from the same basin (Ortiz et al., 2004). Other multi-proxy reconstructions from southern IP span only the Holoceneor part thereof (e.g., Laguna de Zoñar; Martín-Puertas et al.,2008). As a consequence, the selected records are distributedmostly across the northern IP, with the exception of Fuentillejomaar, which is located in central Spain (Table 1, Fig. 1).

The geology of the IP is remarkably diverse, but, in a simplisticway, can be divided into three main geological units (Gibbons andMoreno, 2002), although their exact boundaries are still underdiscussion (Vera, 2004): (1) Palaeozoic and Proterozoic rocksforming the Iberian Massif and the basement of other mountainranges (e.g., Pyrenees); (2) Mesozoic and Cenozoic sedimentaryformations affected by the Alpine orogeny, and mostly constitutingthe Pyrenees, Betics and Iberian Ranges, and (3) large tectonicCenozoic basins, such as the Ebro or Tagus basins and other smallbasins located within the Alpine ranges (Fig. 1). Thus, in northernIberia, the Pyrenees, Cantabrian Cordillera and Galaico-LeonesMountains constitute the most important orographic featureswhile the central IP is crossed by the Central Range, which dividesthe central plateau in two northern and southern “mesetas”. TheIberian Range, which runs north-west to south-east, constitutes thehydrological divide between the Atlantic and Mediterraneanwatersheds (Fig. 1). Due to the geographic situation and topo-graphic conditions, the climate of the IP is extremely varied, butroughly, a moderate Continental climate characterizes the inlandareas, an Oceanic climate dominates in the north and west anda warm Mediterranean climate is experienced along the Mediter-ranean coast (Capel Molina, 1981). Both geography and climate

critically influence the distribution of vegetation and determine thebiogeographical features of all the provinces within the Euro-siberian and Mediterranean regions (Blanco-Castro et al., 1997;Rivas-Martínez, 2007) (see also Fig. 1 in González-Sampériz et al.,in press).

Unfortunately, the large geological, climatic and biogeographicdiversity of the IP is far from being representatively sampled by theselected lake records included in thiswork (Table 1 and Fig.1). Someareas remain poorly covered, such as the central region, due to thelack of multi-proxy studies on the scarce lacustrine systems (cf.Fuentillejo maar; Vegas et al., in press), while other environmentsare over-represented, such as the montane sectors, due to moreabundant permanent, deep lakes, which originated during the lastdeglaciation (e.g., Enol Lake; Moreno et al., in press-a). To coversome of the gaps, other well-known, relatively long records (e.g.,Area Longa in the NW; Gómez-Orellana et al., 2007, or Abric Romaníin the NE, Burjachs and Julià, 1994) are included in the discussiondespite the fact that they do not fulfill the palaeoenvironmentalcriteria established above for site selection since they mainlyconcern vegetation reconstruction. Furthermore, the last glacialcycle is not homogenously represented by the selected records sincelake sequences including MIS 4 or MIS 5aed in the IP are very rare.For these intervals, we support the palaeoclimate discussion withother terrestrial (moraines, speleothems) or marine archives (bothrepresented by black squares in Fig.1). An exhaustive compilation ofpollen records from the IP covering the Pleistocene has beenrecently published by González-Sampériz et al. (in press). In addi-tion, a new issue of Journal of Paleolimnology (Valero-Garcés andMoreno, in press) includes a good compilation of papers based onIberian lake records, though mostly focused on the Holocene.

A. Moreno et al. / Quaternary Science Reviews 36 (2012) 139e153142

3. Methods

An important advance in palaeoclimate reconstruction based onlake records in the IP has been the consistent application of a multi-proxy methodology, following the PAGES strategy and the proce-dure implemented, among others, by the Limnological ResearchCenter from the University of Minnesota (http://lrc.geo.umn.edu).This procedure starts with the Initial Core Description (ICD)including non-destructive measurement of physical properties(usually carried out by a multi-sensor core logging GEOTEK andincluding the measurement of magnetic susceptibility -MS-, bulkdensity, etc.), core splitting into working and archive halves,imaging of the core sections, and macro- and microscopic identi-fication of sedimentary structures and composition using visualand microscopic observations (Schnurrenberger et al., 2001)(Fig. 2). The sedimentological analyses characterize the evolution ofthe depositional environment of the lake and, in combination withother geological and biological data, allow reconstruction of pastclimatic variability (Valero-Garcés et al., 2003) (Fig. 2).

Among the geological proxies, themain palaeoindicators used toidentify and characterize the sedimentary processes controlling theinput, transport and deposition of sedimentary particles, i.e.essential information for understanding the infilling of the lacus-trine system are: (1) mineralogical composition, derived fromX-raydiffraction analyses; (2) elemental geochemistry, obtained at high-resolution by X-ray fluorescence (XRF) core scanning (Last, 2001) oras discrete samples by other methods (ICP, conventional XRF); (3)concentration of total organic (TOC) and inorganic (TIC) carbon, and(4) stable isotope composition (d18O and d13C) in carbonates or bulkorganic matter (Fig. 2). The combined analysis of these proxiesprovides important information regarding, for example, the input

Fig. 2. Flow diagram showing the multi-proxy approach followed in palaeoclimate reconstibility; OM, organic matter; TOC, total organic carbon; TIC, total inorganic carbon; TN, tota

and composition of detrital minerals versus the precipitation ofendogenic components (Corella et al., in press), or data about thehydrological balance and temperature of lake water (Morellónet al., 2009a). Among the biological proxies used for palae-olimnological reconstructions, the most commonly employed are(1) pollen, (2) diatoms, (3) ostracods and/or (4) chironomids (e.g.,Moreno et al., in press-b) (Fig. 2). These indicators provide infor-mation related to the type and extension of the vegetation cover(e.g., Carrión, 2002) and also environmental (temperature, precip-itation) and limnological (pH, lake level, nutrients, water columnmixing) conditions in the lake (e.g., Leira, 2005). The integratedmulti-proxy approach in the study of lake sequences is critical fordisentangling the different forcings influencing lacustrine systems,an indispensable pre-requisite for robust reconstructions ofclimatic variability.

The chronology in the selected records was mainly based on theAMS 14C technique and the dates were calibrated for this reviewusing the INTCAL09 calibration curve (Reimer et al., 2009) (seeSupplementary Table S1). Additionally, other dating techniqueswere used, such as UeTh disintegration series in the carbonatesfrom Banyoles record (Pérez-Obiol and Julià, 1994); OpticallyStimulated Luminescence (OSL) in Villarquemado palaeolake(Valero-Garcés et al., 2007), and palaeomagnetism excursions inFuentillejomaar (Vegas et al., in press). Final construction of the agemodels was carried out by linear interpolation between theobtained dates, except on Enol and Estanya lakes where a general-ised mixed-effect regression was used, following Heegaard et al.(2005). Although the records selected for this review are charac-terized by robust chronological control, some general problems arenevertheless evident (e.g. calibration difficulties for the datesbeyond 45,000 years in longer sequences such as Fuentillejo maar,

tructions from lake sediments (modified from Morellón, 2009). MS, magnetic suscep-l nitrogen; BSi, biogenic silica.

A. Moreno et al. / Quaternary Science Reviews 36 (2012) 139e153 143

scarcity of organic terrestrial remains in glacial lakes such as EnolLake, etc.) that remain difficult to overcome. However, whennecessary, these limitations are discussed in order to avoid misin-terpretation of the main climate trends.

4. The Iberian climate reconstruction during last glacial cycle

Very few multi-proxy studies from lake records in the IP coverthe time interval from last glacial inception (ca. 120 ka) to the“global LGM”.1 In fact, from Table 1 we can only cite Fuentillejomaar (142.4 m) (Vegas et al., in press) and Villarquemado palae-olake (74 m) sequences, both obtained in present-day dry lakesusing a truck-mounted drilling system. Several sequences coverMIS 3 and a larger number includes MIS 2 (Table 1).

4.1. The beginning of last glacial cycle in Northern Iberia (MIS 5 andMIS 4)

The Greenland NGRIP ice core offers an undisturbed record ofthe last glacial inception and reveals a rapid event, D/O 25, occur-ring about 115,000 yrs ago when the northern hemisphere icevolume reached about one third of its glacial extent (NGRIPMembers, 2004). Mediterranean pollen data show that the inter-glacial forest environment is preserved during this period (meanpercentage of temperate pollen around 40e50%) but also respon-ded to rapid D/O events, indicating that the early glacial millennial-scale variability in Greenland has an European counterpart(Tzedakis et al., 2003; Masson-Delmotte et al., 2005; Sánchez-Goñiet al., 2008). In the IP, the full details of the nature and timing of theonset of last glacial cycle and its possible correlation with otherNorth Atlantic marine records and Greenland ice cores are not fullyconstrained. The most detailed available information comes fromIberian margin marine records, which yield information aboutpalaeoceanographic conditions and, through pollen analysis anddirect land-sea correlation, provide evidence of regional-scalevegetation changes during the last glacial inception (e.g., ODP977/A: Martrat et al., 2004; Pérez-Folgado et al., 2004; ODP976:Combourieu Nebout et al., 2002; MD95-2042: Sánchez-Goñi et al.,1999, 2008; MD99-2331: Sánchez-Goñi et al., 2005; MD04-2845:Sánchez-Goñi et al., 2008). These studies indicate a w10� south-ward displacement of vegetation belts inwestern Europe as early asw121 ka as part of continental-scale vegetation changeswhichmayhave played a role in triggering the last glaciation (Sánchez-Goñiet al., 2005). Overall, an apparent synchrony with global climateevents is shown, both in sea surface temperatures (Martrat et al.,2004) and pollen data (Sánchez-Goñi et al., 2008), reflectingmillennial-scale climate variability associated with MIS 5 substagesand D/O events 25e19, and following a long-term trend towardsa cold and arid glacial scenario.

In the terrestrial realm, the lack of well-dated lacustrinesequences for this period prevents the detailed characterization ofthe beginning of last glacial period on land and the nature andimpacts of rapid climate oscillations. As an example, the availablechronology for the Fuentillejo maar record is not yet clear beyondthe limits of the 14C method, except for a magnetic reversal at thebase that provides evidence of the Matuyama-Brunhes boundary

1 We will use the term “global LGM”, according to EPILOG (EnvironmentalProcesses of the Ice Age: Land, Oceans, Glaciers, http://www.glacialoceanatlas.org/index.php?option¼com_content&view¼artilcle&id¼55&itemid¼2) project, for theperiod from 23,000 to 19,000 yrs BP that refers to the time of maximum extent ofthe ice sheets during the last glaciation e the Würm or Wisconsin glaciation (Mixet al., 2001). In Iberian Peninsula, the time of maximum glacier extension does notcorrespond to the global LGM.

(780 ka) (Vegas et al., in press) and extends the record to at least thebeginning of the Middle Pleistocene.

The Villarquemado palaeolake sequence was dated bycombining 14C (for the uppermost 20 m) and OSL (for theremaining 52 m) techniques, yielding a basal age of ca. 120 ka, thuscovering the period from MIS 5 to present-day (Fig. 3). This recordlacks an adequate time control for the interval between 20e48 m(corresponding to w41.5 kae72.5 ka). Thus, boundaries betweenMIS5eMIS4 and MIS4eMIS3 were placed in Fig. 3 on the basis ofsedimentary unit boundaries. The Villarquemado sequence iscomposed of peatbog, alluvial fan and carbonate lake deposits andthe basin was likely a variable mosaic of these three depositionalenvironments during its evolution. In this sense, development ofa carbonate lake (with high contents of Ca and TIC and lower MSvalues) represents higher lake levels than a peatbog setting (higherTOC, lower MS) while alluvial fan deposits (lower carbonate andTOC content, higher MS) represent the lowest lake levels in thebasin. Thus, in the Villarquemado sequence, TOC values are higherduring the Holocene (Unit I, 0e3 m) and MIS 5 (Units VI and VII,37e74 m) (Fig. 3) with the most significant development ofwetlands of the whole sequence, characterized by the alternationof peatbog and shallow carbonate lake environments. A significantdepositional change in the basin is recorded at the onset of MIS 4,with the retreat of the wetlands and the progradation of the distalalluvial fans indicative of a tendency towards lower lake levels(Unit V, 29e37 m, Fig. 3).

Other Iberian records based on pollen data also show largechanges at the onset of the last glacial cycle. In the NW IP, the AreaLonga sequence, recovered from a beach cliff, spans the intervalfromMIS 5c toMIS 3 (Gómez-Orellana et al., 2007) (Fig.1). The baseof this pollen record (ascribed to MIS 5c, corresponding to St.Germain I phase) is dominated by deciduous woodland (Alnus,Quercus robur type, Corylus, Betula and Carpinus) with highproportions of Fagus. During MIS 4, high percentages of Erica, Cal-luna and Poaceae indicate heath and temperate grassland as thepredominant vegetation types with a low abundance of conifersand persistence of meso-thermophytes such us Quercus robur type,Corylus, Fagus, Carpinus, Ulmus and Ilex. The authors’ interpretion isthat while the NW IP was affected by cooling that occurred globallyduring MIS 4, its climate continued to be relatively humid, mostlybased on the high Ericaceae and Poaceae percentages and the lowsteppe taxa values (Artemisia, Chenopodiaceae) that dominate theherbaceous component. In NE Spain, the Abric Romaní travertinerock shelter provides palaeobotanical information for the interval70,000e40,000 years BP (Burjachs and Julià, 1994) (Fig. 1). Treepollen percentages in the oldest deposits (attributed to MIS 5a)reach 40e60%, dominated by pines but with a continuous presenceof Juniperus, Rhamnus, Quercus, Olea-Phillyrea, Betula, Fagus, Pistaciaand other mesothermophilous taxa. The transition to MIS 4 repre-sents a cold but humid phase with less thermophilous taxa(Burjachs and Julià, 1994).

Therefore, up to now and until more data from Villarquemadopalaeolake are available, we can summarize from the scarce avail-able terrestrial records covering MIS 5 to MIS 4 that a consistentclimatic changewas observed across the IP in terms of temperature,with cooling after ca. 65,000 cal years BP. In contrast, patterns ofmoisture availability appear more variable, as detected frommarine pollen data. Thus, records from the northern and north-western margins of the IP indicate cool, humid conditionspromoting the development of Ericaceae and conifers during MIS4 (e.g., MD04-2845 and MD99-2331 marine cores: Sánchez-Goñiet al., 2005, 2008, respectively), while records from the southernmargins indicate drier conditions, with greater development ofsemi-desert vegetation (e.g., MD95-2042 and ODP site 976,reviewed in Fletcher et al., 2010). At all sites, however, a trend of

Fig. 3. Sedimentary sequence for Villarquemado palaeolake record. From left to right: sedimentary units and sedimentological profile, magnetic susceptibility (MS) (in SI units), Ca(in counts per second units) measured by the X-ray Fluorescence (XRF) core scanner, and TIC (total inorganic carbon) and TOC (total organic carbon) percentages. An interpretationof the inferred depositional environments for each unit is presented together with the preliminary chronology (marine isotope stages e MIS e from 5 to 1). Available AMS 14C (inbold type) and eOSL dates (in italics) are shown to the left.

A. Moreno et al. / Quaternary Science Reviews 36 (2012) 139e153144

gradually increasing aridity over the MIS 4 interval is apparent(Sánchez-Goñi et al., 2008; Fletcher et al., 2010).

Glacier records from the Central Pyrenees (García-Ruíz et al.,2003; Pallàs et al., 2006; Lewis et al., 2009) provide coherentsupport for the prevalence of relatively humid conditions at thetransition between MIS 5 and MIS 4 in northern IP. Thus, the mostexternal moraines in the Spanish Central Pyrenees are dated by OSLat 85� 5 ka (Peña et al., 2003; Lewis et al., 2009), placing the timingof the “Iberian last glacial maximum” close to the transitionbetween MIS 5 and MIS 4 (García-Ruíz et al., 2010). This scenario ofcold temperatures, significant humidity across the northern IP, anda gradual decline in humidity across MIS 4, may partly underlinewhy the timing of maximum extent of otherMediterranean glaciersis much earlier than the global LGM (see a review in Hughes andWoodward, 2008). Besides the asynchrony in the maximum iceextent, there is also a discrepancy in the timing of last deglaciation,which appears to have occurred earlier in the Pyrenees (García-Ruízet al., 2003; Pallàs et al., 2006; Lewis et al., 2009) and the Canta-brian mountains (Jiménez Sánchez and Farias, 2002) than in otherEuropean mountains. An explanation for this early glacier retreatmay be found in the abrupt climate changes that occurred later,during MIS 3.

4.2. The record of rapid climate cycles in lake sediments (MIS 3)

Since the study carried out by Lebreiro et al. (1996), where thefirst evidence of Heinrich layers was found in marine sedimentsoffshore Portugal, many other records, mostly from marine cores,have highlighted abrupt fluctuations in the Iberian climate duringMIS 3 synchronous with HE and D/O cycles (e.g. Cacho et al., 1999;Frigola et al., 2008). From the palynological study on marine cores,it is now accepted that those fluctuations also produced importantchanges on land, mostly via changes in water availability andtemperature that could have a great impact on vegetation cover(Sánchez-Goñi et al., 2000, 2002, 2008; Roucoux et al., 2001, 2005;Combourieu Nebout et al., 2002, 2009; Fletcher and Sánchez-Goñi,2008; Naughton et al., 2009; Fletcher et al., 2010). In addition, otherterrestrial tracers measured on marine sediments, such as indica-tors of fluvial and aeolian activity (Moreno et al., 2002, 2005; Bout-Roumazeilles et al., 2007; Frigola et al., 2008), also point tomillennial-scale D/O fluctuations in IP aridity (Fig 4). Recent high-resolution studies detected a two-phase hydrological pattern forsome HE in a marine core offshore Galicia (Naughton et al., 2007)which has been subsequently confirmed by a speleothem recordfrom northern Iberia (Moreno et al., 2010).

A. Moreno et al. / Quaternary Science Reviews 36 (2012) 139e153 145

In contrast to the relatively high number of marine recordscovering this time interval, lake sequences from the IP coveringMIS3 and demonstrating a response on land to rapid climate oscilla-tions are scarce. In fact, even considering lacustrine records ata European scale, the lake sequences where D/O cycles have beenclearly observed and dated are limited (e.g., Allen et al., 1999;Wohlfarth et al., 2008). Considering that lakes are very sensitiveecosystems to small environmental changes, why are MIS 3 climatefluctuations not more clearly recorded? The most plausible expla-nation is that sampling resolution has generally not been highenough, limited in some cases by low glacial sedimentation ratesand compounded by the difficulties of constructing accurate

Fig. 4. Selected marine and terrestrial records from the IP covering GS-2 and GS-3. Fromoffshore Oporto, Portugal (de Abreu et al., 2003); d18O (& VPDB) from El Pindal cave (Morenfrom El Portalet peatbog (González-Sampériz et al., 2006); reconstructed salinity from EstaJulià, 1994; Valero-Garcés et al., 1998); (%) TiO2 from Fuentillejo maar (Vegas et al., in pressinsolation at 65�N; SST (�C) from MD95-2043 record (Cacho et al., 1999) and NGRIP d18O (&moving average (thicker line). DO-I are labelled from 1 to 8. Shaded bands indicated the am2039 and MD95-2040 cores (de Abreu et al., 2003).

chronologies for this time period (i.e., the 14C method is close to itsmaximum limit and, additionally, lake sediments, particularly fromproglacial lakes, are characterized by low organic content duringthis interval thus restricting even more the dating potential(Moreno et al., in press-a)). Although laminated records fromkarstic lakes will probably provide better candidates (with morerobust chronologies supported by counting annual laminae andhigher sedimentation rates permitting the detection of abruptchanges), there is no record in the IP studied up to now with suchfeatures.

In the Villarquemado palaeolake, the aridity trend that startedduring MIS 4 continued and peaked during the lower part of MIS 3

up to down: (%) of N. pachyderma (sinistra) from MD95-2039 and MD95-2040 coreso et al., 2010); Ca (cps) profile from Enol Lake (Moreno et al., in press-a); (%) Juniperusnya Lake (Morellón et al., 2009a); d13C (& VPDB) from Banyoles Lake (Pérez-Obiol and); reconstructed fluvial activity from MD95-2343 record (Frigola et al., 2008); summerVSMOW) record from Greenland (Rasmussen et al., 2006) and smoothed with a 5-pointplitude of HE, positioned following the record of N. pachyderma (sinistra) from MD95-

A. Moreno et al. / Quaternary Science Reviews 36 (2012) 139e153146

(Unit IV, 21e29 m; Fig. 3) where sedimentological evidence forephemeral lake conditions (dolomite formation, red, oxidized finesediments) is present. After around 40,000 cal yrs BP, an alterna-tion of shallow carbonate lake deposits and distal clastic alluvialfan materials reflect rapid hydrological and climate fluctuationsduring MIS 3, although the ascription to individual events is stillnot possible with the available chronological model. More datesthroughout the MIS3 interval and the palynological study of thewhole sequence, currently in progress, will aid the detection ofMIS 3 variability. Although dating uncertainties are high in theFuentillejo maar record from central IP (Table 1, Fig. 1) due tolinear interpolation between very few dates (6 AMS 14C dates forthe last 50,000 years), several fluctuations ascribed to HE andother stadials of the D/O cycles have been identified and inter-preted as arid periods (Vegas et al., in press). Based on thecombination of several proxies (sedimentology, geochemistry,pollen, etc.), HE5 and HE3 have been identified as relatively warmperiods while HE4, 2 and 1 were significantly colder (TiO2percentage is plotted in Fig. 4 as a proxy for dry/cold conditions).The authors refer to regional processes as the cause of modifica-tions in the intensity and persistence of these rapid climateoscillations (Vegas et al., in press).

The site that provided initial clues about MIS 3 climate fluctu-ations in the IP is the Banyoles pollen record, first published byPérez-Obiol and Julià (1994). A later study of sedimentary faciesand stable isotopes on charophytes from the same littoral corereveals impacts on the sediments of HE 3 and 2 that are interpretedas dry periods characterized by lower lake levels (Valero-Garcéset al., 1998) (Fig. 4). Besides Banyoles, other locations in thenorthern IP, notably El Portalet peatbog and Enol Lake (Table 1,Fig. 1), responded to the arid and cold conditions of HE3 and HE2(González-Sampériz et al., 2006; Moreno et al., in press-a) (Fig. 4).Particularly clear is the record of El Portalet peatbog where anincrease in steppe taxa and a decrease in Juniperus frequencies,together with a more abundant siliciclastic component in thesediments, occurred during cold and arid phases associated withrapid events of climate change (González-Sampériz et al., 2006).

Dating the base of sedimentary sequences obtained from pro-glacial lakes or glaciolacustrine deposits has provided usefulinformation for reconstructing the deglaciation stages in theSpanish mountains during MIS 3 (González-Sampériz et al., 2005).There are four noteworthy proglacial lake records that support anearly deglaciation: (1) a basal age of 32.5 ka from El Portalet peat-bog at 1802 m a.s.l. (González-Sampériz et al., 2006); (2) a basal ageof around 33.9 ka from Tramacastilla glacial lake at 1640 m a.s.l.(García-Ruíz et al., 2003; Pallàs et al., 2006; Lewis et al., 2009), bothlocated in the Pyrenees; (3) a basal age of 38 ka from Lago Enol inthe Cantabrian Mountains at 1075 m a.s.l. (Farias-Arquer et al.,1996; Moreno et al., in press-a); and (4) a basal age of 25.5 kafrom Lago de Sanabria in NW Spain at 997 m a.s.l (Rico et al., 2007).All these ages postdate glacier activity in the area and, since thelakes are located at or close to the headwaters of the differentbasins, and behind terminal moraines, it means that the glaciershad already retreated to their cirques or very close to them by40e30 ka.

Although several hypotheses have been postulated, up to nowa satisfactory explanation for the early glacier retreat has not yetbeen found (Gillespie and Molnar, 1995). However, it seems clearthat it was related to the high sensitivity of Mediterranean moun-tain glaciers to climate changes resulting from their distinctivecharacteristics such as their geographical location and their smallersize (Hughes and Woodward, 2008). Recently, García-Ruíz et al.(2010) have proposed that the sustained increase of the Scandi-navian inlandsis between 80 and 55 ka BP (Svendsen et al., 2004)had parallels in the Mediterranean mountains, with rapid glacier

growth that lead to maximum ice extension of some of the glaciertongues approximately at the transition from MIS 5 to MIS 4. Lateron, during MIS 3, and due to the well-known abrupt climate fluc-tuations associated with the D/O cycles, the Scandinavian inlandsismay have stabilized thanks to its larger inertia, but the Mediter-ranean glaciers may have experienced a noticeable retreat duringwarm events. It is interesting to note that the Villarquemado recordalso points to more humid conditions during MIS 4 and MIS 2 thanduring MIS 3 (Fig. 3), coherent with higher long-term moistureavailability in the IP as a pre-requisite for glacier advances.

More records from lakes and glacier evolution and an increasedeffort on dating, possibly combining dating techniques (14C, OSL),are necessary to go further in the identification of the effects onland of rapid climate changes during MIS 3.

4.3. From the global LGM to the Holocene onset (MIS 2/GS-2)

The global LGM can be defined as the most recent interval whenglobal ice sheets reached their maximum integrated volume duringthe last glaciation (Mix et al., 2001). However, as we noted above,the glacier advance associated with the global LGM may be ofsmaller magnitude for Mediterranean, and particularly Iberianglaciers, than that which occurred during MIS 4 (García-Ruíz et al.,2010). The period since the global LGM to the Holocene onset (GS-2,GI-1 and GS-1 in the INTIMATE nomenclature; Lowe et al., 2008) iswell-represented in many marine records surrounding the IP (e.g.,Cacho et al., 2001; Jiménez-Espejo et al., 2007; Naughton et al.,2007; Combourieu Nebout et al., 2009; Fletcher et al., 2010), andit appears as a period with high variability, including events ofabrupt climate change such as HE2 and HE1 and rapid climatefluctuations during LGIT (GI-1, GS-1). Additionally, many Iberianlake records (see Table 1, Figs. 4 and 5) cover this time interval andcan provide some answers to questions about the nature, timing,regional particularities and spatial variability of the main climatechanges in the IP since global LGM.

4.3.1. Was the global LGM the coldest and driest interval of MIS 2 inthe IP?

One of the most important questions to be addressed in relationto climate variability in the IP is the signal on land of the global LGM(GS-2b). Although it is now evident that the global LGM does notcorrespond in most Iberian mountains to the maximum glacierextension (Lewis et al., 2009), was that period the coldest intervalof the last ca. 25,000 years? Was it relatively wet or dry? Marinerecords from Iberian margins indicate that the global LGM,although undoubtedly cold, was not the coldest interval in themarine realm (e.g., Alboran Sea, Cacho et al., 1999; Portuguesemargin, de Abreu et al., 2003) (Fig. 4). In contrast, HE1 (dated about16,000 years BP) is generally marked by the highest percentages ofcold foraminifer Neogloboquadrina pachyderma (s), the highestvalues of IRD, or the lowest SST reconstructed for the last 23,000years. In terms of hydrological changes, HE1 appears also drier thanglobal LGM in offshore Menorca record (based on the K/Al ratio asindicator of fluvial activity in Frigola et al., 2008, see Fig. 4) and inmany marine pollen records (Beaudouin et al., 2007; Naughtonet al., 2007; Combourieu Nebout et al., 2009; Fletcher et al.,2010). Model simulations obtained a clear reduction in bothtemperature of the coldest month and in precipitation for the HE1interval respect to global LGM in Iberia and highlighted a moresignificant response on the European Atlantic coast that decreasesvery rapidly inland (Kageyama et al., 2005). Data from continentalsequences in the IP, related to temperature and water availabilitycomparing global LGM and HE1, are available to corroborate orreject those model outputs.

Fig. 5. Selected marine and terrestrial records from the IP covering from 18,000 to 8000 cal yrs BP. From up to down: (%) of N. pachyderma (sinistra) from MD95-2039 offshoreOporto, Portugal (de Abreu et al., 2003); d18O (& VPDB) from El Pindal cave (Moreno et al., 2010); (%) Juniperus from El Portalet peatbog (González-Sampériz et al., 2006);reconstructed salinity from Estanya Lake (Morellón et al., 2009a); broad tendencies of d13C (& VPDB) from Banyoles Lake (Pérez-Obiol and Julià, 1994; Valero-Garcés et al., 1998); (%)TiO2 from Fuentillejo maar (Vegas et al., in press); summer insolation at 65�N; SST (�C) from MD95-2043 record (Cacho et al., 1999) and NGRIP d18O (& VSMOW) record fromGreenland (Rasmussen et al., 2006) and smoothed with a 5-point moving average (thicker line). Shaded bands indicated the amplitude of short abrupt events during deglaciationand arrows mark tendencies (see text for discussion).

A. Moreno et al. / Quaternary Science Reviews 36 (2012) 139e153 147

In general, recently studied lake sequences from the IP supportprevious interpretations from marine sediments, and in particularare in agreement with the relatively humid hydrological signal ofglobal LGM. In Villarquemado palaeolake (Figs. 1 and 3), MIS 2 ischaracterized by a decrease in alluvial fan activity and more

development of carbonate lake environments than before, pointingto relatively humid conditions during the LGM. In Estanya Lake(Fig. 1, Morellón et al., 2009b), a shallow carbonate-producing lakesystem during the global LGM (from the onset of the lake sequence,ca. 21,000e18,000 cal yrs BP), contrasts with a closed, permanent

A. Moreno et al. / Quaternary Science Reviews 36 (2012) 139e153148

saline lake characterized by an evaporitic dominant sedimentation(starting at 18,000 and lasting until 14,000 cal yrs BP) (Fig. 4).Therefore, the global LGM was not the driest interval in the Pre-Pyrenees and the significant reduction in runoff occurred after-wards (Morellón et al., 2009a). Additionally, the preservation oflacustrine sediments in several records from playa-lakes in theCentral Ebro Basin during the global LGM (see summary inGonzalez-Samperiz et al., 2008), suggests phases of increasedmoisture during this period. Thus, the global LGM was probablycharacterized by periods of positive hydrological balance perhapscaused by reduced summer insolation at the latitude of Iberia(Fig. 4). If that was the case, evapotranspiration during the summermonths may have decreased, contributing to relatively high lakelevels without a significant increase in rainfall, as suggested by thereconstruction provided by the Estanya Lake record (Morellón et al.,2009a). An additional factor with the potential to increase wateravailability in certain areas is the expected high fluvial dischargeproduced in relation to the deglaciation process in the mountains(Valero-Garcés et al., 2004; González-Sampériz et al., 2005) whichhad already started by this time. There is some evidence of thatprocess in the form of flood deposits in global LGM terracesindicative of a period of high discharge (Sancho-Marcén et al.,2003) that correlates with an increase in fluvial activity just afterglobal LGM (Frigola et al., 2008).

Although temperatures are usually more difficult to reconstructfrom lake sediments than hydrological balance (Cohen, 2003),pollen data from lacustrine sequences provide clear evidence forthe IP of a cold scenario for the global LGM until the beginning ofthe Bølling/Allerød (see compilation in González-Sampériz et al., inpress): the landscape was dominated by cold steppe formationswith a minor presence of conifers and restricted occurrence ofmeso-thermophytes. In sequences with higher sample andtemporal resolution, detailed interpretation of pollen spectraprovides evidence for a particularly cold interval associated withHE1. Thus, in El Portalet peatbog, HE1 is detected by the presence ofgray siliciclastic silts indicating low lake productivity, a decrease inJuniperus and increase of steppe taxa (González-Sampériz et al.,2006). Similarly, more positive values of d13C in carbonates werefound in Banyoles record (Pérez-Obiol and Julià, 1994; Valero-Garcés et al., 1998) (Fig. 4).

From all the recent evidence outlined above, we can concludethat the most arid and coldest period in the IP during GS-2occurred in within the GS-2a (Fig. 4). This interval has been calledthe “Mystery Interval” (MI) (Denton et al., 2005), and embraces themarine HE1 thus corresponding to the first phase of last glacialtermination (17.5e14.5 cal kyr BP). In the Enol Lake record, the MIcorresponds to the lowest linear sedimentation rate of the wholesequence pointing to very low runoff and thus little transport tothe lake (Moreno et al., in press-a). In addition, the MI coincideswith a hiatus in the formation of a speleothem from El Pindal Cave,in northern Spain, also suggesting a dry (and cold) period (Morenoet al., 2010). The same stalagmite grew during the global LGM,pointing to less extreme climate conditions at that time comparedto the MI (Fig. 4). However, up to now, the evidence from lakes (orspeleothems) has not been sufficiently accurate to discriminatechronologically whether the arid period includes the whole MIinterval (ca. GS-2a) or whether it is more constrained to HE1, asseems to be the case from marine temperature records (Cachoet al., 2001). In fact, some sequences record two pulses duringthe MI (e.g., Estanya Lake salinity reconstruction or Fuentillejomaar TIO2 aridity indicator) while others (e.g., Juniperus percent-ages in El Portalet peatbog) only point to one longer cold/dry eventembracing the whole GS-2a interval (Fig. 4).

Despite chronological uncertainties and the different responsessuggested by the available lake records (i.e. one or two pulses), the

important effect of the Meridional Overturning Circulation (MOC)on the IP climate and the rapid response of terrestrial ecosystems toMOC variability is evident. The MI marks the start of the first phaseof the last glacial termination (T1a) and was characterized by thestrong reduction of MOC (McManus et al., 2004) in comparison toLGM levels due to high rates of freshwater input during icebergdischarges of HE1. The shutdown in MOC lasted 2000 yr and causedextremely cold winter temperatures in the North Atlantic area(Denton et al., 2005) and likely formed sea ice, reduced sea surfaceevaporation and consequently produced dry conditions in Europe(Wohlfarth et al., 2008) and into Asia (Cheng et al., 2006). There-fore, as a consequence of the close connection between westernEuropean temperatures and MOC intensity, IP temperatures arecolder during the MI than during the earlier global LGM period.

4.3.2. When and how did the last deglaciation occur in the IP?Terminology for the last deglaciation was first defined from the

Fennoscandian region based on pollen sequences and the corre-sponding vegetation changes, including periods such as theBølling-Allerød or the Younger Dryas (Mangerud et al., 1974), thatcorrespond in the INTIMATE nomenclature referring to Greenlandice records toGI-1 andGS-1, respectively (Björck et al.,1998) (Fig. 5).The last deglaciationwas characterized by a series of abrupt climaticchanges (GI-1aeGI-1e, GS-1), with broadly similar trends identifiedin palaeoclimate records obtained from many sites throughout theNorth Atlantic region. However, the extent to which the NorthAtlantic sequence of climatic changes is reflected in palaeoclimaticrecords from the IP, in terms of timing and pattern of the abruptclimatic changes, is still a matter of debate (e.g., Carrión et al., inpress). From marine cores surrounding the IP, at least two particu-larities with respect to Greenland records have arisen: (1) theearliest onset of warming associated with the first phase of the lastdeglaciation occurred atw15.5 cal kyr BP, prior to further andmoremarked warming at the onset of the GI-1 (Fletcher et al., 2010), and(2) a stable e to warming trend in sea surface temperatures duringGI-1 is observed in contrast to the cooling trend recorded inGreenland (Cacho et al., 2001). Furthermore, recent analyses ofpollen records in southern Iberianmarine cores indicate short-livedintervals of forest decline consistent with cooling and drying duringthe GI-1d (Older Dryas) and GI-1b (Inter-Allerød Cold Period)(Combourieu Nebout et al., 2009; Fletcher et al., 2010). The lack ofaccurate chronologies and high-resolution analyses in continentalrecords has precluded the identification of abrupt climate changeswithinGI-1until recently (e.g., González-Sampériz et al., 2006).Newlake sequences like Villarquemado palaeolake, combining the studyof vegetation and the response of the lake system itself to climatechanges, will provide key information for the characterization ofabrupt changes experienced during last deglaciation.

In the northeastern IP, the hydrological response to abruptclimate change during the last deglaciation has been described inEstanya Lake (Fig. 1). In this record, the onset of GI-1 is detected bychanges in sedimentation in the lake and a significant negativeexcursion of d13Corg values reflecting an increase in organicproductivity likely related to deeper lake level conditions (Morellónet al., 2009a). The salinity reconstruction also points to a morepositive hydrological balance during GI-1 and shows minorchanges in response to short abrupt cold events, such as GI-1d andGI-1b, pointing to slightly drier conditions (Fig. 5). Similarly, themontane peatbog record from El Portalet reflects a decline inherbaceous steppe association, typical of glacial conditions, and anexpansion of pioneer deciduous trees at the beginning of GI-1.Vegetation cover and sediment composition also reacted rapidly toshorter cold events with the deposition of siliciclastic silts and anincrease in steppe plants and a decrease in Juniperus (González-Sampériz et al., 2006) (Fig. 5).

A. Moreno et al. / Quaternary Science Reviews 36 (2012) 139e153 149

In Laguna Grande and Laguna del Hornillo, both located in thewestern Iberian Range, the sequence of events within GI-1 havebeen identified by characterizing the laminations and the type andcontent of organic matter (Vegas, 2006). In these two lakes, anarid and cold event (GI-1d) is found between GI-1a (Allerød) andGI-1e (Bølling) but there is no signal around GI-1b, probably dueto the low temporal resolution of the record. In other lakes fromthe wider Mediterranean region, a similar hydrological response toGI-1d and GI-1b events is observed (e.g., Lago dell’Accesa inCentral Italy; Magny et al., 2006). On the contrary, an opposedpalaeohydrological pattern is observed in central-western Europe,where G1-1d and GI-1b are characterized by higher lake levels inthe Swiss Plateau, Jura mountains and French Pre-Alps (Magny,2001). This latitudinal division in the hydrological responseduring abrupt climate changes occurring throughout last deglaci-ation, has been recently explained by the prevalence of “blockingepisodes” that will favor or prevent cyclone penetration into theMediterranean or northern and central Europe (Fletcher et al.,2010).

In the available lake records (Fig. 5), the onset of the warmingtrend associated with the Bølling period is synchronous, within agemodel uncertainties, with the onset of GI-1 in Greenland, but thepattern observed is more gradual than abrupt. Additionally, inEstanya Lake record, the Allerød period appears wetter than theBølling period, in a similar way to that recorded in El Pindal cavelocated in northwestern Spain (Moreno et al., 2010) (Fig. 5). Simi-larly, the El Portalet pollen record reflects a generally reducedpresence of steppe taxa and the first development, as opposed tooccasional presence, of Corylus during the Allerød in contrast to theBølling (González-Sampériz et al., 2006). Therefore, this pattern isconsistent with Mediterranean marine SST records (Cacho et al.,2001) and differs from Greenland ice record where warmertemperatures over Greenland were reached abruptly at the onset ofthe Bølling period and declined afterwards (Fig. 5). The similarresponse of some lake (González-Sampériz et al., 2006; Morellónet al., 2009a) and speleothem (Moreno et al., 2010) sequencesfrom northern IP and Mediterranean marine SST records (Cachoet al., 2001) to the global warming related to the first phase ofthe last glacial termination 1 (T1a), reflects a particular reaction interms of temperature and water availability of this southernEuropean region. This pattern may relate to a continental-scaleNeS latitudinal pattern of changing climatic evolution over the GI-1interval as proposed by Genty et al. (2006), which should be bettercharacterized for the IP with future studies.

4.3.3. Timing, synchrony and ecosystems response to the YoungerDryas and the Holocene onset

The second phase of last glacial termination (T1b) correspondsto the secondweakening of theMOC during the Younger Dryas coldperiod, probably also triggered by a discharge of glacial meltwater(Hughen et al., 2000; McManus et al., 2004). While a clear responseduring the GS-1 interval (or Younger Dryas, YD) is detected inmarine environments of the Iberian margin, mostly in terms ofreduced sea surface temperatures (e.g. Cacho et al., 2001), clearresponse is less evident in continental archives from the IberianPeninsula where a variable vegetation response is observeddepending on the altitude and latitude of the studied records(Carrión et al., in press). Thus, changes in the landscape and vege-tation cover during the YD appear to be more marked in moun-tainous areas (e.g., El Portalet peatbog record indicates that the lakewas frozen all-year round, González-Sampériz et al., 2006) than inmid-to-low altitude sites (e.g., Lake Banyoles; Pérez-Obiol and Julià,1994).

Other indicators measured in lake sequences besides vegeta-tion are plotted in Fig. 5 and their combination supports the

existence of a YD event in the northern IP as a dry and coldperiod without clear geographical variability. Thus, a lake leveldrop and salinity increase in Estanya Lake were indicated by thereturn to deposition of gypsum-rich facies and an abruptdecrease in organic productivity (marked by positive excursion ofd13Corg and a sharp decrease in Bio Si) (Morellón et al., 2009a). InBanyoles Lake, the isotopic composition of authigenic carbonates(d18O and d13C) reaches peak values at around 12,000 years(Valero-Garcés et al., 1998) while sedimentation in El Portaletdecreased dramatically or even ceased during the GS-1 inresponse to the previously mentioned permanent freezing of thelake (González-Sampériz et al., 2006). In Enol Lake, gray silici-clastic silts with low organic content and pollen spectra domi-nated by herbaceous taxa characterize an open landscape withscarce vegetation during the GS-1 unit (Moreno et al., in press-b).Similarly, the presence of massive clayey silts with low organiccontent in the Fuentillejo maar record (Vegas et al., in press), andsignificant changes in sediment stratigraphy and diatoms asso-ciation in the Laguna Grande at Sierra de Neila (Vegas et al.,2003), indicate a cold and arid climate associated with the GS-1 interval.

Thus, considering high and low altitude sites, the response toGS-1 in the IP lake records seems identical (Fig. 5). This finding mayindicate that the different signals to the same climatic eventrecorded in the pollen spectra from different IP regions was linkedto the distance to vegetation refuges that controlled the timing andintensity of the vegetation response. In addition, since most of thecases that are considered to show an “unexpected” response to GS-1 lie in the Mediterranean-influenced climate region (Fig. 1),a centennial to millennial-scale resilience of the established forestscan be presented as another explanation to account for the differentvegetation responses (Gil-Romera et al., 2010; Carrión et al., inpress). This view, however, is not in agreement with the findingsof palynological research on Mediterranean marine cores, whichsuggest a rapid response of the Mediterranean forest cover tocentennial-scale variability, both at the abrupt onset of the YD andwithin the GS-1 interval (Combourieu Nebout et al., 2009; Fletcheret al., 2010).

The onset of the Holocene represents an abrupt climate changetowards warmer and, in general, wetter climates at 11,600 cal yrsBP (e.g., Hoek et al., 2008). Although this transition was apparentlysynchronous in different records from the IP, optimum Holoceneclimate conditions were not reached at the same time (Morellónet al., 2009a). In Estanya Lake, sedimentary and geochemicalproxies indicate that the lowest lake level of the whole sequence(last 20,000 years) occurred from 11,600 to 9400 cal yrs BP, whenfull Holocene conditions were finally reached (Morellón et al.,2009a). The Lake Banyoles sequence also records the eventualdecrease in steppe taxa at 9500 cal yrs BP (Pérez-Obiol and Julià,1994). In Enol Lake record, wetter conditions were not founduntil 9800 cal yrs BP when Ca, TOC and TIC percentages increasewhile siliciclastic particles decrease (Moreno et al., in press-b). Inthat record, arboreal pollen values increasemarkedly at the onset ofthe Holocene, dominated by a rapid increase of deciduous Quercus(45%), although the highest values were recorded at 9700 cal yrs BP.Accordingly, pollen records from the Alboran Sea indicate that thetemperate Mediterranean forest expanded dramatically inresponse to increased humidity not developed at the Holoceneonset but at 10,600 cal yrs BP (Fletcher et al., 2010). This delay maybe related to a restricted rainy season during the boreal summerinsolation maximum (Tzedakis, 2007). Thus, it seems from theavailable records, that the delay in the Holocene onset is relatedmore to hydrological parameters than to temperature changes,pointing to a possible impact of the monsoon dynamics on the IPclimate.

A. Moreno et al. / Quaternary Science Reviews 36 (2012) 139e153150

5. Summary and ideas for the future work

Selected lake records show the IP response to abrupt climatechanges during last glacial cycle. Although, in general, there isa synchrony and a high correlation with North Atlantic regionclimate, the IP presents some peculiarities likely related to itssouthern location and the mix of African and European influenceson its climate. Thus, the transition from MIS 5 to MIS 4 appears asa cold but relatively wet period, and corresponds to the maximumglacier extension in the northern Iberian mountains (e.g., Pyrenees,Cantabrian Mountains). Subsequent deglaciation occurs rapidly,probably associated with the general tendency towards greateraridity during MIS 4, and due to abrupt climate changes thatcharacterized the MIS 3 interval, which includes some of the mostarid periods in Iberian continental records. Abrupt climate changes,particularly HE, are observed in several records by changes in thesediment and vegetation cover and composition, thus demon-strating the effect of rapid climate variability on land. The globalLGM is not the coldest or the most arid interval of the last 25,000years since the MI, and the embedded HE1 event, are characterizedby the highest aridity in the studied sequences. As detected in thelake sequences, the Lateglacial period starts synchronously totemperature increase in Greenland (14,600 cal yrs BP), but thepattern is not so abrupt and, additionally, the highest humidity isreached at the end of GI-1 (Allerød) and not at the beginning(Bølling). Finally, the GS-1 (YD) is observed in the hydrologicalresponse of the lake records but variable signals in the pollenspectra, suggesting different sensitivity of the vegetation indifferent localities with respect to altitude, topography and micro-climate, and possibly relating to vegetation resilience at this time.The Holocene climatic optimum in terms of humidity seems to bedelayed with respect to other European records, being reached indifferent locations only after 10.5e9.5 cal yrs BP.

From this compilation, it is evident that a major advance hasbeen achieved recently in terms of palaeoclimate reconstructionsobtained from lake records in the IP. Many of the records thatprovide critical information have been published recently or are inpress (Estanya Lake, Enol Lake, etc.). However, despite the increasednumber of new studies, several questions remain open due to thelack of high-resolution records in key geographic regions. Thus, thesouthern IP region was not extensively discussed in this paper dueto the scarcity of multi-proxy high-resolution lake records. It isclear that more records are necessary, especially from low altitudeareas, that are currently underrepresented in the compilation. Thegreatest effort must be made to obtain laminated records, e.g., inkarstic lakes such as Banyoles Lake, that will provide better reso-lution permitting the detection and characterization of abruptclimate changes during the last glacial cycle. In addition, longsequences such as Villarquemado palaeolake will provide newinformation on climate changes during the last glacial inceptionand the IP LGM. It is strongly advisable to compare and combineinformation from lake records with those obtained from othercontinental palaeoarchives, particularly speleothems and glacialdeposits, and terrestrial tracers in marine sediment sequences. Theintegration of data from different palaeoarchives is critical todeveloping the understanding of the response of continental Iberiato rapid climate changes during last glacial cycle.

The multi-proxy approach has been found to be the best (if notonly!) option to discriminate climate changes from other morelocal influences on the lake records (particular response of vege-tation, etc.). However, further efforts are required not only tocombine indicators, but to improve their calibration with theinstrumental record. Greater use of quantitative estimations oftemperature and precipitation would be highly informative andthis remains an under-explored approach in the IP. Proxy

calibration, together with an improvement of transfer functiondatabases, will lead to better reconstruction of climate signals andwill thus also contribute to the improvement of climate models.

Finally, the construction of robust chronological frameworks isindispensable for palaeoclimate reconstruction, particularly for thecharacterization of rapid climate changes. More effort must bemade to look for high-quality dating material (terrestrial macro-remains, charcoal) suitable for 14C AMS in lake sediments. Inaddition, other methods, such as the tephrochronology, have notbeen explored in the IP terrestrial records and may be worth tryingdespite the non-favourable situationwith respect to major volcaniczones and prevailing wind directions. Comparing records withindependent chronologies (i.e., not tuned respect to Greenland icecores) is essential for the identification of leads and lags in thecontinental response to different climate events.

Acknowledgements

The funding for this study mainly derives from LIMNOCLIBER(REN2003-09130-C02-02), LIMNOCAL (CGL2006-13327-C04-01),GRACCIE-CONSOLIDER (CSD2007-00067) and DINAMO (CGL2009-07992) projects, provided by the Spanish Inter-Ministry Commis-sion of Science and Technology (CICYT), and the VILLARQUEMADO(P196/2005) project from the Aragón Regional Government (DGA).A. Moreno acknowledges the funding from the “Ramón y Cajal”postdoctoral program. We are grateful to Ma Paz Errea for his helpwith Fig 1 and to Juana Vegas (IGME, Madrid, Spain), Mayte Rico(IPE-CSIC, Zaragoza, Spain), Lucia de Abreu (Cambridge University,UK), and Jaume Frigola and Isabel Cacho (UB, Barcelona, Spain) forkindly providing their data. We are indebt to the organizers ofINTIMATEmeeting in Oxford (September 2008) where the ideas forthis study originated.

Appendix. Supplementary material

Supplementary data associated with this article can be found inthe online version, at doi:10.1016/j.quascirev.2010.06.031.

References

de Abreu, L., Shackleton, N.J., Schönfeld, J., Hall, M.A., Chapman, M.R., 2003.Millennial-scale oceanic climate variability off the Western Iberian marginduring the last two glacial periods. Marine Geology 196, 1e20.

Allen, J.R.M., Brandt, U., Brauer, A., Hubberten, H.W., Huntley, B., Keller, J., Kraml, M.,Mackensen, A., Mingram, J., Negendank, J.F.W., Nowaczyk, N.R., Oberhänsli, H.,Watts, W.A., Wulf, S., Zolitschka, B., 1999. Rapid environmental changes insouthern Europe during the last glacial period. Nature 400, 740e743.

Beaudouin, C., Jouet, G., Suc, J.P., Berné, S., Escarguel, G., 2007. Vegetation dynamicsin southern France during the last 30 ky BP in the light of marine palynology.Quaternary Science Reviews 26, 1037e1054.

Björck, S., Walker, M.J.C., Cwynar, L., Johnsen, S.J., Knudsen, K.-L., Lowe, J.J.,Wohlfarth, B., INTIMATE Group, 1998. An event sratigraphy for the last termi-nation in the North Atlantic based on the Greenland Ice core record: a proposalby the INTIMATE group. Journal of Quaternary Science 13, 283e292.

Blanco-Castro, E., Casado, M., Costa, M., Escribano, R., García-Antón, M., Génova, M.,Gómez, A., Moreno, J., Morla, C., Regato, P., Sainz Ollero, H., 1997. Los bosquesibéricos. Una interpretación geobotánica. Planeta, Barcelona.

Bout-Roumazeilles, V., Combourieu Nebout, N., Peyron, O., Cortijo, E., Landais, A.,Masson-Delmotte, V., 2007. Connection between South Mediterranean climateand North African atmospheric circulation during the last 50,000 yr BP NorthAtlantic cold events. Quaternary Science Reviews 26, 3197e3215.

Broecker, W.S., 2000. Abrupt climate change: causal constraints provided by thepaleoclimate record. Earth-Science Reviews 51, 137e154.

Burjachs, F., Julià, R., 1994. Abrupt climatic changes during the last glaciation basedon pollen analysis of the Abric Romani, Catalonia, Spain. Quaternary Research42, 308e315.

Cabrera, L., Anadón, P., 2003. Desarrollo de los proyectos IGCP 219 y 324 (GLOPALS-GGLAB) en España: Kerry Kelts como elemento dinamizador. In: Valero-Garcés, B. (Ed.), Limnogeology in Spain: A Tribute to Kerry Kelts. ConsejoSuperior de Investigaciones Científicas, Madrid, pp. 37e59.

Cacho, I., Grimalt, J.O., Canals, M., Sbaffi, L., Shackleton, N.J., Schönfeld, J., Zahn, R.,2001. Variability of the Western Mediterranean sea surface temperatures

A. Moreno et al. / Quaternary Science Reviews 36 (2012) 139e153 151

during the last 25,000 years and its connection with the northern hemisphereclimatic changes. Paleoceanography 16, 40e52.

Cacho, I., Grimalt, J.O., Pelejero, C., Canals, M., Sierro, F.J., Flores, J.A., Shackleton, N.J.,1999. Dansgaard-Oeschger and Heinrich event imprints in Alboran Seatemperatures. Paleoceanography 14, 698e705.

Capel Molina, J.J., 1981. Los climas de España. Oikos-tau ediciones, Barcelona.Carrión, J.S., 2002. Patterns and processes of Late Quaternary environmental change

in a montane region of southwestern Europe. Quaternary Science Reviews 21,2047e2066.

Carrión, J.S., Fernández, S., González-Sampériz, P., Gil-Romera, G., Badal, E., Carrión-Marco, Y., López-Merino, L., López-Sáez, J.A., Burjachs, F., Expected trends andsurprises in the Lateglacial and Holocene vegetation history of the IberianPeninsula and Balearic Islands. Review of Palaeobotany and Palynology, in press(doi:10.1016/j.revpalbo.2009.12.007).

Clark, P.U., Dyke, A.S., Shakun, J.D., Carlson, A.E., Clark, J., Wohlfarth, B.,Mitrovica, J.X., Hostetler, S.W., McCabe, A.M., 2009. The last glacial maximum.Science 325, 710e714.

Cohen, A.S., 2003. Paleolimnology. The History and Evolution of Lake Systems.Oxford University Press, New York.

Combourieu Nebout, N., Peyron, O., Dormoy, I., Desprat, S., Beaudouin, C.,Kotthoff, U., Marret, F., 2009. Rapid climatic variability in the west Mediterra-nean during the last 25 000 years from high resolution pollen data. Climate ofthe Past 5, 503e521.

Combourieu Nebout, N., Turon, J.L., Zahn, R., Capotondi, L., Londeix, L., Pahnke, K.,2002. Enhanced aridity and atmospheric high-pressure stability over thewestern Mediterranean during the North Atlantic cold events of the past 50 k.y.Geology 30, 863e866.

Corella, J.P., Valero-Garcés, B.L., Moreno, A., Morellón, M., Rull, V., Giralt, S., Rico,M.T., Pérez-Sanz, A., Sedimentary evolution and palaeohydrology of karstic,meromictic Montcortés Lake (Spanish Pre-Pyrenees) during the last 6,000years. Journal of Paleolimnology, in press (doi:10.1007/s10933-010-9443-3).

Costanza, R., Graumlich, L., Steffen, W., Crumley, C., Dearing, J., Hibbard, K.,Leemans, R., Redman, C., Schimel, D., 2007. Sustainability or collapse: what canwe learn from integrating the history of humans and the rest of nature? Ambio36, 522e527.

Cheng, H., Edwards, R.L., Wang, Y., Kong, X., Ming, Y., Kelly, M.J., Wang, X.,Gallup, C.D., 2006. A penultimate glacial monsoon record from Hulu Cave andtwo-phase glacial terminations. Geology 34, 217e220.

Denton, G.H., Alley, R.B., Comer, G.C., Broecker, W.S., 2005. The role of seasonality inabrupt climate change. Quaternary Science Reviews 24, 1159e1182.

Denton, G.H., Broecker, W., Alley, R.B., 2006. The mystery interval 17.5 to 14.5 kyrago. In: Brigham-Grette, J., Kull, C., Kiefer, T. (Eds.), PAGES News, pp. 14e16.

Farias-Arquer, P., Jiménez Sánchez, M., Martínez, J., 1996. Nuevos datos sobre laestratigrafía del relleno cuaternario de la depresión de Comeya, Picos deEuropa, Asturias. Geogaceta 20, 2016e2019.

Fletcher, W.J., Sánchez-Goñi, M.F., 2008. Orbital- and sub-orbital-scale climateimpacts on vegetation of the western Mediterranean basin over the last 48,000yr. Quaternary Research 70, 451e464.

Fletcher, W.J., Sánchez-Goñi, M.F., Allen, J.R.M., Cheddadi, R., CombourieuNebout, N., Huntley, B., Lawson, I., Londeix, L., Magri, D., Margari, V., Müller, U.C.,Naughton, F., Novenko, E., Roucoux, K., Tzedakis, P.C., 2010. Millennial-scalevariability during the last glacial in vegetation records from Europe. QuaternaryScience Reviews 29, 2839e2864.

Fletcher, W.J., Sánchez-Goñi, M.F., Peyron, O., Dormoy, I., 2010. Abrupt climatechanges of the last deglaciation detected in a western Mediterranean forestrecord. Climate of the Past 6, 245e264.

Frigola, J., Moreno, A., Cacho, I., Canals, M., Sierro, F.J., Flores, J.A., Grimalt, J.O., 2008.Evidence of abrupt changes in Western Mediterranean Deep Water circulationduring the last 50 kyr: a high-resolution marine record from the Balearic Sea.Quaternary International 181, 88e104.

Fritz, S.C., 2008. Deciphering climate history from lake sediments. Journal of Pale-olimnology 39, 5e16.

García-Ruíz, J.M., Moreno, A., González-Sampériz, P., Valero-Garcés, B., Martí-Bono, C., 2010. La cronología del último ciclo glaciar en las montañas del sur deEuropa. Una revisión. Cuaternario y Geomorfología. 24, 35e46.

García-Ruíz, J.M., Valero-Garcés, B.L., Martí-Bono, C., González-Sampériz, P., 2003.Asynchroneity of maximum glacier advances in the central Spanish Pyrenees.Journal of Quaternary Science 18, 61e72.

Genty, D., Blamart, D., Ghaleb, B., Plagnes, V., Causse, Ch., Bakalowicz, M., Zouari, K.,Chkir, N., Hellstrom, J., Wainer, K., Bourges, F., 2006. Timing and dynamics of thelast deglaciation from European and North African d13C stalagmite profi-lesdcomparison with Chinese and South Hemisphere stalagmites. QuaternaryScience Reviews 25, 2118e2142.

Gibbons, W., Moreno, T., 2002. The Geology of Spain. The Geological Society,London.

Gil-Romera, G., López-Merino, L., Carrión, J.S., González-Sampériz, P., Martín-Puertas, C., Sáez, J.A.L., Fernández, S., Antón, M.G., Stefanova, V., 2010. Inter-preting resilience through long-term ecology: potential insights in westernMediterranean landscapes. The Open Ecology Journal 2010 (3), 43e53.

Gillespie, A., Molnar, P., 1995. Asynchronous maximum advances of mountain andcontinental glaciers. Reviews of Geophysics 33, 311e364.

Gómez-Orellana, L., Ramil-Rego, P., Muñoz Sobrino, C., 2007. The Würm in NWIberia, a pollen record from Area Longa (Galicia). Quaternary Research 67,438e452.

González-Sampériz, P., Leroy, S.A.G., Fernández, S., García-Antón, M., Gil-García,M.J., Uzquiano, P., Valero-Garcés, B., Figueiral, I., Steppes, savannahs, forests andphytodiversity reservoirs during the Pleistocene in the Iberian Peninsula.Review of Palaeobotany and Palynology, in press (doi:10.1016/j.revpalbo.2010.03.009).

González-Sampériz, P., Valero-Garcés, B.L., Carrión, J.S., Peña-Monné, J.L., García-Ruíz, J.M., Martí-Bono, C., 2005. Glacial and Lateglacial vegetation in north-eastern Spain: new data and a review. Quaternary International 140e141,4e20.

González-Sampériz, P., Valero-Garcés, B.L., Moreno, A., Jalut, G., García-Ruiz, J.M.,Martí-Bono, C., Delgado-Huertas, A., Navas, A., Otto, T., Dedoubat, J.J., 2006.Climate variability in the Spanish Pyrenees during the last 30,000 yr revealedby the El Portalet sequence. Quaternary Research 66, 38e52.

Gonzalez-Samperiz, P., Valero-Garces, B.L., Moreno, A., Morellon, M., Navas, A.,Machin, J., Delgado-Huertas, A., 2008. Vegetation changes and hydrologicalfluctuations in the Central Ebro Basin (NE Spain) since the Late Glacial period:saline lake records. Palaeogeography, Palaeoclimatology, Palaeoecology 259,157e181.

Heegaard, E., Birks, H.J.B., Telford, R.J., 2005. Relationships between calibrated agesand depth in stratigraphical sequences: an estimation procedure by mixed-effect regression. The Holocene 15, 612e618.

Hoek, W.Z., Yu, Z.C., Lowe, J.J., 2008. INTegration of Ice-core, MArine, and TErrestrialrecords (INTIMATE): refining the record of the Last GlacialeInterglacial tran-sition. Quaternary Science Reviews 27, 1e5.

Hughen, K.A., Southon, J.R., Lehman, S.J., Overpeck, J.T., 2000. Synchronousradiocarbon and climate shifts during the last deglaciation. Science 290,1951e1954.

Hughes, P.D., Woodward, J.C., 2008. Timing of glaciation in the Mediterraneanmountains during the last cold stage. Journal of Quaternary Science 23,575e588.

Jiménez-Espejo, F.J., Martínez-Ruiz, F., Sakamoto, T., Ijima, K., Gallego-Torres, D.,Harada, N., 2007. Paleoenvironmental changes in the western Mediterraneansince the last glacial maximum: high resolution multiproxy record from theAlgero-Balearic basin. Palaeogeography, Palaeoclimatology, Palaeoecology 246,292e306.

Jiménez Sánchez, M., Farias, P., 2002. New radiometric and geomorphologicevidences of a last glacial maximum older than 18 ka in SW European moun-tains: the example of Redes Natural Park (Cantabrian Mountains, NW Spain).Geodinamica Acta 15, 93e101.

Kageyama, M., Combourieu Nebout, N., Sepulchre, P., Peyron, O., Krinner, G.,Ramstein, G., Cazet, J.-P., 2005. The Last Glacial Maximum and Heinrich Event 1in terms of climate and vegetation around the Alboran Sea: a preliminarymodel-data comparison. Comptes Rendues a l’Acadèmie des Sciences de Paris337, 983e992.

Lebreiro, S.M., Moreno, J.C., McCave, I.N., Weaver, P.P.E., 1996. Evidence for Heinrichlayers off Portugal (Tore Seamount: 39�N, 12�W). Marine Geology 131, 47e56.

Leira, M., 2005. Diatom responses to Holocene environmental changes in a smalllake in northwest Spain. Quaternary International 140e141, 90e102.

Lewis, C.J., McDonald, E.V., Sancho, C., Peña, J.L., Rhodes, E.J., 2009. Climatic impli-cations of correlated Upper Pleistocene glacial and fluvial deposits on the Cincaand Gállego Rivers (NE Spain) based on OSL dating and soil stratigraphy. Globaland Planetary Change 67, 141e152.

Lowe, J.J., Rasmussen, S.O., Björck, S., Hoek, W.Z., Steffensen, J.P., Walker, M.J.C.,Yu, Z.C., 2008. Synchronisation of palaeoenvironmental events in the NorthAtlantic region during the last termination: a revised protocol recommended bythe INTIMATE group. Quaternary Science Reviews 27, 6e17.

Magny, M., 2001. Palaeohydrological changes as reflected by lake-level fluctuationsin the Swiss Plateau, the Jura Mountains and the northern French Pre-Alpsduring the Last GlacialeHolocene transition: a regional synthesis. Global andPlanetary Change 30, 85e101.

Magny, M., de Beaulieu, J.-L., Drescher-Schneider, R., Vannière, B., Walter-Simonnet, A.-V., Millet, L., Bossuet, G., Peyron, O., 2006. Climatic oscillations incentral Italy during the Last GlacialeHolocene transition: the record from LakeAccesa. Journal of Quaternary Science 21, 311e320.

Mangerud, J., Andersen, S.T., Berglund, B.E., Donner, J.J., 1974. Quaternary stratig-raphy of Norden, a proposal for terminology and classification. Boreas 4,109e128.

Martín-Puertas, C., Valero-Garcés, B.L., Mata, P., González-Sampériz, P., Bao, R.,Moreno, A., Stefanova, V., 2008. Arid and humid phases in Southern Spainduring the last 4000 years: the Zoñar Lake record, Córdoba. The Holocene 40,195e215.

Martrat, B., Grimalt, J.O., Lopez-Martinez, C., Cacho, I., Sierro, F.J., Flores, J.A.,Zahn, R., Canals, M., Curtis, J.H., Hodell, D.A., 2004. Abrupt temperature changesin the Western Mediterranean over the past 250000 years. Science 306,1762e1765.

Masson-Delmotte, V., Landais, A., Combourieu Nebout, N., von Grafenstein, U.,Jouzel, J., Caillon, N., Chappellaz, J., Dahl-Jensen, D., Johnsen, S.J., Stenni, B.,2005. Rapid climate variability during warm and cold periods in polar regionsand Europe. Comptes Rendus Geosciences 337, 935e946.

McManus, J., Francois, R., Gherardi, J.M., Keigwin, L., Brown-Leger, S., 2004. Collapseand rapid resumption of Atlantic meridional circulation linked to deglacialclimate changes. Nature 428, 834e837.

Mix, A.C., Bard, E., Schneider, R., 2001. Environmental processes of the ice age: land,oceans, glaciers (EPILOG). Quaternary Science Reviews 20, 627e657.

A. Moreno et al. / Quaternary Science Reviews 36 (2012) 139e153152

Montserrat, J., 1992. Evolución glaciar y postglaciar del clima y la vegetación en lavertiente sur del Pirineo: estudio palinológico. Instituto Pirenaico de Ecología,Zaragoza.

Morellón, M., 2009. Paleohydrology and Abrupt Climate Changes in NortheasternIberian Peninsula During the last 20,000 Years: The Lacustrine Record ofEstanya (Huesca). Earth Sciences Department, University of Zaragoza, Zaragoza,p. 323.

Morellón, M., Valero-Garcés, B., González-Sampériz, P., Vegas-Vilarrúbia, T., Rubio,E., Rieradevall, M., Delgado-Huertas, A., Mata, P., Romero, Ó., Engstrom, D.R.,López-Vicente, M., Navas, A., Soto, J., Climate and human signatures during theMedieval Warm Period and the Little Ice Age in the Spanish Pre-Pyrenees: theLake Estanya record. Journal of Paleolimnology, in press (doi:10.1007/s10933-009-9346-3).

Morellón, M., Valero-Garcés, B., Vegas-Vilarrúbia, T., González-Sampériz, P.,Romero, Ó., Delgado-Huertas, A., Mata, P., Moreno, A., Rico, M., Corella, J.P.,2009a. Lateglacial and Holocene palaeohydrology in the western Mediterra-nean region: the Lake Estanya record (NE Spain). Quaternary Science Reviews28, 2582e2599.

Morellón, M., Valero-Garcés, B.L., Anselmetti, F., Ariztegui, D., Schnellmann, M.,Moreno, A., Mata, P., Rico, M., Corella, J.P., 2009b. Late Quaternary depositionand facies model for karstic Lake Estanya (NE Spain). Sedimentology 56,1505e1534.

Moreno, A., Cacho, I., Canals, M., Grimalt, J.O., Sánchez-Goñi, M.F., Shackleton, N.J.,Sierro, F.J., 2005. Links between marine and atmospheric processes oscillatingat millennial time-scale. A multy-proxy study of the last 50,000 yr from theAlboran Sea (Western Mediterranean Sea). Quaternary Science Reviews 24,1623e1636.

Moreno, A., Cacho, I., Canals, M., Prins, M.A., Sánchez-Goñi, M.F., Grimalt, J.O.,Weltje, G.J., 2002. Saharan dust transport and high latitude glacial climaticvariability: the Alboran Sea record. Quaternary Research 58, 318e328.

Moreno, A., Stoll, H.M., Jiménez Sánchez, M., Cacho, I., Valero-Garcés, B., Ito, E.,Edwards, L.R., 2010. A speleothem record of rapid climatic shifts during lastglacial period from Northern Iberian Peninsula. Global and Planetary Change 71,218e231. doi:10.1016/j.gloplacha.2009.1010.1002.

Moreno, A., Valero-Garcés, B.L., Jiménez Sánchez, M., Domínguez, M.J., Mata, P.,Navas, A., González-Sampériz, P., Stoll, H., Farias, P., Morellón, M., Corella, P.,Rico, M., The last deglaciation in the Picos de Europa National Park (CantabrianMountains, Northern Spain). Journal of Quaternary Science, in press-a (doi:10.1002/jqs.1265).

Moreno, A., López-Merino, L., Leira, M., Marco-Barba, J., González-Sampériz, P.,Valero-Garcés, B., López-Sáez, J.A., Santos, L., Mata, P., Ito, E., Revealing the last13,500 years of environmental history from the multiproxy record of a moun-tain lake (Lago Enol, northern Iberian Peninsula). Journal of Paleolimnology, inpress-b (doi:10.1007/s10933-10009-19387-10937).

Naughton, F., Sánchez-Goñi, M.F., Desprat, S., Turon, J.L., Duprat, J., Malaize, B.,Joli, C., Cortijo, E., Drago, T., Freitas, M.C., 2007. Present-day and past (last 25000years) marine pollen signal off western Iberia. Marine Micropaleontology 62,91e114.

Naughton, F., Sánchez-Goñi, M.F., Kageyama, M., Bard, E., Duprat, J., Cortijo, E.,Desprat, S., Malaizé, B., Joly, C., Rostek, F., Turon, J.L., 2009. Wet to dry climatictrend in north-western Iberia within Heinrich events. Earth and PlanetaryScience Letters 284, 329e342.

NGRIP Members, 2004. High-resolution record of Northern Hemisphere climateextending into the last interglacial period. Nature 431, 147e151.

Ortiz, J., Torres, T., Delgado-Huertas, A., Julià, R., Lucini, M., Llamas, F.J., Reyes, E.,Soler, V., Valle, M., 2004. The palaeoenvironmental and palaeohydrologicalevolution of Padul Peat Bog (Granada, Spain) over one million years, fromelemental, isotopic and molecular organic geochemical proxies. OrganicGeochemistry 35, 1243e1260.

Pallàs, R., Rodés, Á., Braucher, R., Carcaillet, J., Ortuño, M., Bordonau, J., Bourlès, D.,Vilaplana, J., Masana, E., Santanach, P., 2006. Late Pleistocene and Holoceneglaciation in the Pyrenees: a critical review and new evidence from 10Beexposure ages, south-central Pyrenees. Quaternary Science Reviews 25,2937e2963.

Peña, J.L., Sancho, C., Lewis, C., McDonald, E., Rhodes, E., 2003. Las morrenas ter-minales de los valles glaciares del Gállego y Cinca (Pirineo de Huesca). Datoscronológicos. Boletín Glaciológico Aragonés 4, 91e109.

Pérez-Folgado, M., Sierro, F.J., Flores, J.A., Grimalt, J.O., Zahn, R., 2004. Paleo-climatic variations in foraminifer assemblages from the Alboran Sea (WesternMediterranean) during the last 150 ka in ODP Site 977. Marine Geology 212,113e131.

Pérez-Obiol, R., Julià, R., 1994. Climate change on the Iberian Peninsula recorded ina 30.000 yr pollen record from Lake Banyoles. Quaternary Research 41, 91e98.

Pons, A., Reille, M., 1988. The Holocene and upper Pleistocene pollen record fromPadul (Granada, Spain): a new study. Palaeogeography, Palaeoclimatology,Palaeoecology 66, 243e263.

Rasmussen, S.O., Andersen, K.K., Svensson, A., Steffensen, J.P., Vinther, B.M., Clausen,H.B., Siggaard-Andersen, M.L., Johnsen, S.J., Larsen, L.B., Dahl-Jensen, D., Bigler,M., Röthlisberger, R., Fisher, H., Goto-Azuma, K., Hansson, M., Ruth, U., 2006. Anew Greenland ice core chronology for the last glacial termination. Journal ofGeophysical Research 11 (doi:10.1029/2005JD006079).

Reimer, P., Baillie, M., Bard, E., Bayliss, A., Beck, J., Blackwell, P., Bronk Ramsey, C.,Buck, C., Burr, G., Edwards, R., Friedrich, M., Grootes, P., Guilderson, T., Hajdas, I.,Heaton, T., Hogg, A., Hughen, K., Kaiser, K., Kromer, B., McCormac, F.,Manning, S., Reimer, R., Richards, D., Southon, J., Talamo, S., Turney, C., van der

Plicht, J., Weyhenmeyer, C., 2009. IntCal09 and Marine09 radiocarbon agecalibration curves, 0e50,000 years cal BP. Radiocarbon 51, 1111e1150.

Rico, M., Garcés, B.V., Vega, J.C., Moreno, A., González-Sampériz, P., Morellón, M.,Mata, P., 2007. El registro del Lago de Sanabria desde la última deglaciación, XIIReunión Nacional de Cuaternario. In: Lario, J., Silva, P.G. (Eds.). AEQUA, Ávila, pp.213e214.

Rivas-Martínez, S., 2007. Mapa de series, geoseries y geopermaseries de vegetaciónde España. Memoria del mapa de la vegetación potencial de España. ItinerariaGeobotanica 17, 1e436.

Roucoux, K.H., de Abreu, L., Shackleton, N.J., Tzedakis, C., 2005. The response of NWIberian vegetation to North Atlantic climate oscillations during the last 65 kyr.Quaternary Science Reviews 24, 1637e1653.

Roucoux, K.H., Shackleton, N.J., de Abreu, L., Schönfeld, J., Tzedakis, C., 2001.Combined marine proxy and pollen analyses reveal rapid Iberian vegetationresponse to North Atlantic millennial-scale climate oscillations. QuaternaryResearch 56, 128e132.

Rull, V., González-Sampériz, P., Valero-Garcés, B., Corella, P., Morellón, M., Vegeta-tion changes in the southern Pyrenean flank during the last millennium inrelation to climate and human activities: the Montcortès lacustrine record.Journal of Paleolimnology, in press (doi:10.1007/s10933-010-9444-2).

Sánchez-Goñi, M.F., Turon, J.L., Eynaud, F., Gendreau, S., 2000. European climaticresponse to millenial-scale changes in the atmosphere-ocean system during theLast Glacial period. Quaternary Research 54, 394e403.

Sánchez-Goñi, M.F., Cacho, I., Turon, J.L., Guiot, J., Sierro, F.J., Peypouquet, J.-P.,Grimalt, J.O., Shackleton, N.J., 2002. Synchroneity between marine and terres-trial responses to millennial scale climatic variability during the last glacialperiod in the Mediterranean region. Climate Dynamics 19, 95e105.

Sánchez-Goñi, M.F., Loutre, M.F., Crucifix, M., Peyron, O., Santos, L., Duprat, J.,Malaizé, B., Turon, J.L., Peypouquet, J.P., 2005. Increasing vegetation and climategradient in Western Europe over the Last Glacial Inception (122e110 ka): data-model comparison. Earth and Planetary Science Letters 231 (1, 2), 111e130.

Sánchez-Goñi, M.F., Landais, A., Fletcher, W.J., Naughton, F., Desprat, S., Duprat, J.,2008. Contrasting impacts of Dansgaard-Oeschger events over a westernEuropean latitudinal transect modulated by orbital parameters. QuaternaryScience Reviews 27, 1136e1151.

Sancho-Marcén, C., Peña-Monné, J.L., Lewis, C., McDonald, E., Rhodes, E., 2003.Preliminary dating of glacial and fluvial deposits in the Cinca river Valley (NESpain): chronological evidences for the Glacial Maximum in the Pyrenees? In:Ruiz Zapata, B., Dorado-Valiño, M., Valdeomillos, A., Gil-García, M.J., Bardají, T.,Bustamante, I., Mendizábal, I. (Eds.), Quaternary Climatic Changes and Envi-ronmental Crises in The Mediterranean Region. Alcalá de Henares, Madrid, pp.169e174.

Schnurrenberger, D.W., Kelts, K., Johnson, T.C., Shane, L.C.K., Ito, E., 2001. Nationallacustrine core repository (LacCore). Journal of Paleolimnology 25, 123e127.

Solomon, S., Qin, D., Manning, M., Chen, Z., Marquis, M., Averyt, K.B., Tignor, M.,Miller, H.L., 2007. Climate change 2007: the physical science basis. Contributionof Working Group I to the Fourth Assessment Report of the IntergovernmentalPanel on Climate Change. Cambridge University Press, Cambridge, UK.

Svendsen, J.I., Alexanderson, H., Astakhov, V.I., Demidov, I., Dowdeswell, J.A.,Funder, S., Gataullin, V., Henriksen, M., Hjort, C., Houmark-Nielsen, M.,Hubberten, H.W., Ingólfsson, Ó., Jakobsson, M., Kjær, K.H., Larsen, E.,Lokrantz, H., Lunkka, J.P., Lyså, A., Mangerud, J., Matiouchkov, A., Murray, A.,Möller, P., Niessen, F., Nikolskaya, O., Polyak, L., Saarnisto, M., Siegert, C.,Siegert, M.J., Spielhagen, R.F., Stein, R., 2004. Late Quaternary ice sheet history ofnorthern Eurasia. Quaternary Science Reviews 23, 1229e1271.

Tzedakis, C., McManus, J., Hooghiemstra, H., Oppo, D.W., Wijmstra, T.A., 2003.Comparison of changes in vegetation in northeast Greece with records ofclimate variability on orbital and suborbital frequencies over the last 450000years. Earth and Planetary Science Letters 212, 197e212.

Tzedakis, P.C., 2007. Seven ambiguities in the Mediterranean palaeoenvironmentalnarrative. Quaternary Science Reviews 26, 2042e2066.

Valero-Garcés, B., González-Sampériz, P., Navas, A., Machín, J., Delgado-Huertas, A.,Peña-Monné, J.L., Sancho-Marcén, C., Stevenson, T., Davis, B., 2004. Palae-ohydrological fluctuations and steppe vegetation during the Last GlacialMaximum in the central Ebro valley (NE Spain). Quaternary International 122,43e55.

Valero-Garcés, B., González-Sampériz, P., Morellón, M., Rico, M., Moreno, A., Navas,A., Machín, J., Mata, M.P., Rubio, J.C., 2007. Climate and Tectonics for the last 100kyrs in the Iberian Range (NE Spain) based on the Villarquemado LacustrineRecord. Fourth International Limnogeology Congress, Barcelona, Spain.

Valero-Garcés, B., Moreno, A., Iberian lacustrine records: responses to past andrecent global changes in the Mediterranean region. Journal of PaleolimnologySpecial Volume, in press.

Valero-Garcés, B., Navas, A., Mata, P., Delgado-Huertas, A., Machín, J., González-Sampériz, P., Moreno, A., Schwalb, A., Ariztegui, D., Schnellmann, M., Bao, R.,González-Barrios, A., 2003. Sedimentary facies analyses in lacustrine cores:from initial core descriptions to detailed palaeoenvironmental reconstructions.A case study from Zoñar Lake (Cordoba province, Spain). In: Valero-Garcés, B.(Ed.), Limnogeology in Spain: A Tribute to Kerry Kelts. C.S.I.C., Madrid, pp.385e414.

Valero-Garcés, B.L., 2003. Limnologeología en España: un tributo a Kerry Kelts/Limnogeology in Spain: a tribute to Kerry Kelts. Biblioteca de Ciencias. ConsejoSuperior de Investigaciones Científicas, Madrid, p. 438.

Valero-Garcés, B.L., Zeroual, E., Kelts, K., 1998. Arid phases in the western Medi-terranean region during the last glacial cycle reconstructed from lacustrine

A. Moreno et al. / Quaternary Science Reviews 36 (2012) 139e153 153

records. In: Benito, G., Baker, V.R., Gregory, K.J. (Eds.), Paleohydrology andEnvironmental Change, pp. 67e80.

Vegas, J., 2006. Los sistemas lacustres de las Sierras de Neila y Urbión: análisissedimentológico y climático del Pleistoceno Superior y Holoceno. Departa-mento de Petrología y Geoquímica. Universidad Complutense de Madrid,Madrid, p. 387.

Vegas, J., 2007. Caracterización de eventos climáticos del Pleistoceno superior-Holoceno mediante el estudio sedimentológico de la Laguna Grande (SierraNeila, NO Sistema Ibérico). Revista de la Sociedad Geológica de España 20,53e70.

Vegas, J., Blanca Ruiz Zapata, M., Ortiz, J.E., Galán, L., Torres, T., García-Cortés, A., Gil-García, M.J., Pérez-González, A., Gallardo-Millán, J.L., Identification of aridphases during the last 50 ca. ka BP from Fuentillejo maar-lacustrine record(Campo de Calatrava Volcanic Field, Spain). Journal of Quaternary Science, inpress (doi:10.1002/jqs.1262).

Vegas, J., Pérez-González, A., Ruiz Zapata, B., Gil-García, M.J., Dorado, M.,Valdeolmillos, A., López-García, M.J., 2003. The GS-1/Younger Dryas Event inthe Laguna Grande lacustrine record. LateglacialeHolocene transition in theNW Iberian Range, Spain. In: Valero-Garcés, B. (Ed.), Limnogeology in Spain: ATribute to Kerry Kelts. Biblioteca de Ciencias, CSIC, Madrid, pp. 284e304.

Vera, J.A., 2004. Geología de España. Instituto Geológico y Minero, Madrid, España,p. 884.

Wohlfarth, B., Veres, D., Ampel, L., Lacourse, T., Blaauw, M., Preusser, F., Andrieu-Ponel, V., Kéravis, D., Lallier-Vergès, E., Björck, S., Davies, S.M., de Beaulieu, J.-L.,Risberg, J., Hormes, A., Kasper, H.U., Possnert, G., Reille, M., Thouveny, N.,Zander, A., 2008. Rapid ecosystem response to abrupt climate changes duringthe last glacial period in western Europe, 40e16 kyr. Geology 36, 407e410.

Wolff, E.W., Chappellaz, J., Blunier, T., Rasmussen, S.O., Svensson, A., 2010. Millen-nial-scale variability during the last glacial: The ice core record. QuaternaryScience Reviews 29, 2828e2838.